Fibre Optics Part 3

Testing Issues

‘If you can’t measure it, you can’t improve it’ Lord Kelvin

There has been crude data available on the amount of fibre in foods since the late nineteenth century, but it has never been easy to accurately measure exactly how much fibre is contained in different items. Clearly, this is not helped by the shifting sands of fibre definitions outlined in Part 2, but it is also tricky because the methods of testing keep on changing.

In the 1890s, fibre generally referred to whatever was left when foods were boiled in acid and then alkali, supposedly mimicking digestion, although this method was only really used to classify animal feeds, rather than through any concern for the dietary health of humans. In fact, for a large part of the twentieth century, fibre was not really considered something humans should concern themselves with when it came to diet, with most healthy eating advice of that period focusing on getting enough calories, protein and micronutrients. Higher fibre foods such as whole grain breads and brown rice were, for the most part, considered low grade and less refined, poor substandard choices for less affluent consumers.

Some Cheap, Low Grade Foods

It was not until the mid 1970s that potential benefits of fibre consumption started to be suggested, particularly a possible impact on cholesterol sequestration by water soluble fibres. Dramatic increases in coronary heart disease, particularly in the US during the 1950s and 60s, were being linked to high blood cholesterol, and evidence was emerging that fibre might be of benefit in managing the condition. As a result, new testing methods were developed to measure the fibre content of foods more accurately, and by the mid-1980s a process known as the Enzymatic Gravimetric Method (EGF) was adopted by the Association of Analytical Chemists.

EGF involves serial treatments of foods with various digestive enzymes, including amylase, amyloglucosidase and protease, to replicate digestion in the human mouth, stomach and small intestine. After treatment, what is left should be stripped of all fat, protein and digestible carbohydrates, and this residue is analysed by a gravimetric testing method to determine the mass of fibre in the original food sample.

A scientist doing some science-type things

By the early 1990s, evidence was building up that fibre had a positive role in cardiovascular health and weight management, and EGF was widely adopted by regulatory authorities around the world as the standard testing method. Fibre rich diets became the prevailing recommendations in developed countries, as they struggled with increasing rates of obesity, CVD and type 2 diabetes. Evidence that fibre may also be beneficial in reducing the risk of colorectal cancers only increased the focus.

For some time, fibres were classified as soluble or insoluble, based on the suggestion that this would have a significant impact on how they were likely to act in the human body. However, it is now clear that these measurements of solubility, based entirely on laboratory testing in water, do not always corelate with how substances behave in the human gut. This has led to the revised classifications outlined in section 1, which in turn led to the discovery of a big discrepancy in EGF testing.

EGF does not pick up resistant starch or non-digestible oligosaccharides, two of the four categories of fibre, even though these are known to resist digestion and impact the microbiome. Because of this discrepancy, in 2009, EGF protocols were amended to cover these compounds, creating a modified process known as mEGF. In many foods, EGF and mEGF produce near identical results, but in ingredients such as lentils, bananas and pulses, there can be considerable differences.

Some Lentils, avoiding being tested using mEGF.

Unfortunately, however, mEGF is a complex, expensive and difficult process, and very few foods have been analysed in this way. Because of the the large number of steps involved, small errors at each stage can build up, leading to a high degree of uncertainty, with errors levels of around 20-30%. This makes the value of mEGF testing extremely limited. Within the food industry, the old EGF testing method was already notorious for being inaccurate, with testing from identical samples frequently showing wildly different values. Making a case for a less accurate, more expensive version has proved difficult.

Often, the figures for fibre shown on nutrition labels are hugely different to those shown in standard food composition tables, and the actual amount of fibre in many common foods is not accurately known. Nearly a decade and a half after the introduction of mEGF, most service laboratories only use the original EGF method, although when test results come back, it is sometimes not completely clear which method has been used. If we want consumers to eat more resistant starch or non-digestible oligosaccharides, there is currently very little good quality evidence regarding where people can find them.

Another issue with current testing is that it is often unclear how products have been handled, with very little standardisation of sample preparation techniques across the industry. Samples might be ground, freeze dried, milled, or otherwise processed as part of the testing procedure. When simple chemical composition is being assessed, these processes are unlikely to have much of an impact. But it is highly probable that these sorts of physical treatments will impact on cell wall structures, altering levels of resistant starch, and so not giving a clear picture of the overall fibre composition. To date, sample preparation techniques are not written into the standard procedures for EGF or mEGF, creating another level of uncertainty.

And even without all this inbuilt inaccuracy, none of the current methods provide much useful information on which types of fibre are contained within food products. Even less detail is available on the cell wall structures, something which can have a huge impact on how our bodies process foods. More detailed analysis is possible, involving techniques such as dissolution kinetics, molecular weight measurement, or analysis of cell wall porosity, but these are expensive and difficult, meaning that this information is almost non-existent in analysis tables for common food products.

In truth, the science of fibre testing is yet to catch up with the state of knowledge, and as such, the amount of fibre in composition tables and on pack does not give a complete picture of how healthy a food product is. Fibre is more complex than other nutrients, and none of the commonly used chemical or physical analysis techniques provide a complete picture. There have been some attempts to make this clearer, with the Glycemic Index being a significant but imperfect example. But the reality is that the fibre picture of most foods is opaque at best, making some items look worse than they are, and giving others an undeserved health halo.

This is an issue. Over the years, it has almost certainly led us towards a less healthy food system, as the benefits of changing or reducing processing techniques are often not clear or easy to communicate. Pure numbers will tell you that a highly processed grain or pulse acts no differently in the body to an intact whole grain or lentil. In fact, the more highly processed option with a sprinkling of added fibre might wrongly appear to be the healthier option.

Processed foods are widely demonised, something that is largely unfair and unjustified. But because of the way we test and categorise foods, the potential health benefits of lower levels of processing are often not completely clear, something that the industry must take some responsibility for. Much of this is due to the nature of the testing methods used, but more broadly, it is down to the insatiable human desire to use numbers, targets and measurements to define health. To fully understand and categorise the healthfulness of food products, it will be necessary to create better, more accurate and more comprehensive testing methods. But to actively encourage people to eat those products will probably require something else entirely new.